VersatileMotion: A Unified Framework for Motion Synthesis and Comprehension

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Ling, Zeyu, Han, Bo, Li, Shiyang, Cheng, Jikang, Shen, Hongdeng, Zou, Changqing
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912393852878848
author Ling, Zeyu
Han, Bo
Li, Shiyang
Cheng, Jikang
Shen, Hongdeng
Zou, Changqing
author_facet Ling, Zeyu
Han, Bo
Li, Shiyang
Cheng, Jikang
Shen, Hongdeng
Zou, Changqing
contents Large language models (LLMs) are, by design, inherently capable of multi-task learning: through a unified next-token prediction paradigm, they can naturally address a wide variety of downstream tasks. Prior work in the motion domain has demonstrated some generality by adapting LLMs via a Motion Tokenizer coupled with an autoregressive Transformer to generate and understand human motion. However, this generality remains limited in scope and yields only modest performance gains. We introduce VersatileMotion, a unified multimodal motion LLM that combines a novel motion tokenizer, integrating VQ-VAE with flow matching, and an autoregressive transformer backbone to seamlessly support at least nine distinct motion-related tasks. VersatileMotion is the first method to handle single-agent and multi-agent motions in a single framework and enable cross-modal conversion between motion, text, music, and speech, achieving state-of-the-art performance on seven of these tasks. Each sequence in MotionHub may include one or more of the following annotations: natural-language captions, music or audio clips, speech transcripts, and multi-agent interaction data. To facilitate evaluation, we define and release benchmark splits covering nine core tasks. Extensive experiments demonstrate the superior performance, versatility, and potential of VersatileMotion as a foundational model for future understanding and generation of motion.
format Preprint
id arxiv_https___arxiv_org_abs_2411_17335
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle VersatileMotion: A Unified Framework for Motion Synthesis and Comprehension
Ling, Zeyu
Han, Bo
Li, Shiyang
Cheng, Jikang
Shen, Hongdeng
Zou, Changqing
Computer Vision and Pattern Recognition
Large language models (LLMs) are, by design, inherently capable of multi-task learning: through a unified next-token prediction paradigm, they can naturally address a wide variety of downstream tasks. Prior work in the motion domain has demonstrated some generality by adapting LLMs via a Motion Tokenizer coupled with an autoregressive Transformer to generate and understand human motion. However, this generality remains limited in scope and yields only modest performance gains. We introduce VersatileMotion, a unified multimodal motion LLM that combines a novel motion tokenizer, integrating VQ-VAE with flow matching, and an autoregressive transformer backbone to seamlessly support at least nine distinct motion-related tasks. VersatileMotion is the first method to handle single-agent and multi-agent motions in a single framework and enable cross-modal conversion between motion, text, music, and speech, achieving state-of-the-art performance on seven of these tasks. Each sequence in MotionHub may include one or more of the following annotations: natural-language captions, music or audio clips, speech transcripts, and multi-agent interaction data. To facilitate evaluation, we define and release benchmark splits covering nine core tasks. Extensive experiments demonstrate the superior performance, versatility, and potential of VersatileMotion as a foundational model for future understanding and generation of motion.
title VersatileMotion: A Unified Framework for Motion Synthesis and Comprehension
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2411.17335